Figure - available from: Pattern Analysis and Applications
This content is subject to copyright. Terms and conditions apply.
Comparison of the average execution time (in microseconds) required for a single pair-wise biometric comparison, boxplots depict the corresponding median, upper/lower quartile and upper/lower whisker for different methods (note the logarithmic scale on x-axis)
Source publication
Scale-invariant feature transform (SIFT), which represents a general purpose image descriptor, has been extensively used in the field of biometric recognition. Focusing on iris biometrics, numerous SIFT-based schemes have been presented in past years, offering an alternative approach to traditional iris recognition, which are designed to extract di...
Similar publications
A biometric system allows an individual to be automatic identification using a distinguishing or single feature possessed by the person. The biometric system of identification available which is regarded as the most accurate and reliable known is iris recognition. In this paper, we discuss the strategies used to construct an Iris Recognition System...
In a cancelable iris recognition technique, all enrollment patterns are masked using a transformation function, and the invertibility process for obtaining the original data should not be possible. A novel cancelable iris authentication approach in the encrypted domain is presented in this paper. The double random phase encoding (DRPE) algorithm in...
Iris recognition is one of the automated processes of verifying individuals' identity based on their iris characteristics. Apparently, the random nature of the iris texture, which is unique for each individual, makes it an exclusive trait for biometric recognition even for the case of identical twins' authentication. Recently, the improvement in de...
Abstract—Iris recognition is one of the automated processes
of verifying individuals’ identity based on their iris
characteristics. Apparently, the random nature of the iris
texture, which is unique for each individual, makes it an
exclusive trait for biometric recognition even for the case of
identical twins’ authentication. Recently, the improvem...
An efficient multimodal biometric system which combines biometric data originated from face, iris and signature biometrics has been presented. Proposed feature extraction algorithm for unimodal and multimodal system has been based on discrete wavelet transform. Among the various biometrics face and iris based human authentication system are proved...
Citations
... When combined with point descriptors such as SIFT (Scale-Invariant Feature Transform), SURF (Speeded Up Robust Features), ORB (Oriented Fast and Rotated BRIEF) [8,9,10,11], Siamese networks can create a highly accurate iris recognition system. The point descriptors are responsible for detecting characteristic points on the image and generating vectors describing these points. ...
... One advantage of this approach is that point descriptors can be used to recognize the iris, even if it is not fully visible in the image. Additionally, Siamese networks with point descriptors have the ability to generalize and can operate with high efficiency on training and test data from different sources [8,9,10,11]. ...
... The Brute Force method uses the Euclidean distance between two descriptors. A smaller distance d v indicates greater similarity between two points (8). ...
One of the most important steps in the operation of biometric systems based on iris recognition of the human eye is pattern comparison. However, the comparison of the recorded pattern with the pattern stored in the database of the biometric system cannot function properly without effective extraction of key features from the iris image. In the presented work, we propose an iris recognition system based on image feature extraction and extreme grey shade analysis. Harris-Laplace, RANSAC and SIFT descriptor algorithms were used to find and describe key points. In the experimental part, two methods were used to compare descriptors: the Brute Force method and the Siamese Network method. IIT Delhi Iris Database (version 1.0), MMU v2 database, UBIRIS v1, UBIRIS v2 image databases were used for the study. The proposed method utilizes a different approach when using the generalized corner extraction algorithm (Harris-Laplace algorithms) for comparing iris patterns. In addition, we prove that the use of the descriptor and the Siamese neural networks significantly improves the results obtained in the original method based on paths alone in the case of well contrasted infrared images with very low resolutions.
... SIFT (Scale-invariant feature transform), which is a scale-invariant feature transform algorithm, is a local feature detection method based on spatial scale extreme points. As the algorithm has invariance to rotation operation, scaling operation and brightness change in addition to strong robustness to noise and the characteristic of scalability, it can accurately extract the corner features in the image [11]. The extracted image features are then matched with the pre-stored charging port features to complete the identification function of the charging port. ...
Automatic charging for electric vehicles has broad development prospects for meeting the personalized service experience of users while overcoming the inherent safety hazards. An identification and positioning approach suitable for engineering applications is the key to promoting automatic charging. In this paper, a low-cost, high-precision method to identify and position charging ports based on SIFT and SGBM is proposed. The feature extraction approach based on SIFT is adopted to produce the difference of Gaussian (DOG) for scale space construction, and the feature matching algorithm with nearest-neighbor search, which is a kind of machine learning, is utilized to yield the map set of matching points. In addition, the disparity calculation is conducted with a semi-global matching algorithm to obtain high-precision positioning results for the charging port. In order to verify the feasibility of the method, a complete identification and positioning experiment of charging port was carried out based on OpenCV and MATLAB.
... Rathgeb et al. (2018) proposed a method for iris recognition by scale invariant transform feature, representing a general purpose image descriptor, discriminative orientation based feature selection, and magnitude possibility distribution function. The weight assignment for the iris texture sub-regions showed an improved performance with equal error rates of 0.88%, and 0.9% for CASIAv3 and MMU, respectively [4]. Divya and Urmila (2016) used Daugman's method to determine the pupil and the iris borders. ...
Iris recognition occupies an important rank among the biometric types of approaches as a result of its accuracy and efficiency. The aim of this paper is to suggest a developed system for iris identification based on the fusion of scale invariant feature transforms (SIFT) along with local binary patterns of features extraction. Several steps have been applied. Firstly, any image type was converted to grayscale. Secondly, localization of the iris was achieved using circular Hough transform. Thirdly, the normalization to convert the polar value to Cartesian using Daugman’s rubber sheet models, followed by histogram equalization to enhance the iris region. Finally, the features were extracted by utilizing the scale invariant feature transformation and local binary pattern. Some sigma and threshold values were used for feature extraction, which achieved the highest rate of recognition. The programming was implemented by using MATLAB 2013. The matching was performed by applying the city block distance. The iris recognition system was built with the use of iris images for 30 individuals in the CASIA v4. 0 database. Every individual has 20 captures for left and right, with a total of 600 pictures. The main findings showed that the values of recognition rates in the proposed system are 98.67% for left eyes and 96.66% for right eyes, among thirty subjects.
... Afterwards, based on the dogman model, various traditional iris recognition system was designed. The traditional iris recognition techniques are broadly classified into three folds: pre-processing (iris segmentation) [11,96,104], feature extraction [32,73,78,150,152,156 and matching/classification [6,64,98,99,127,[209][210][211]. Iris recognition system affected by the images taken from the different sensor device. ...
... However, the method was mathematically complex. To improve storage requirement and authentication speed, Rathgeb et al. [152] proposed an improved SIFT based Iris recognition system. However, the proposed method is a computational complex. ...
... A deep learning-based method is robust to acquire feature precisely from the raw image, without the involvement of a handcrafted feature, which significantly enhances the matching method [41,60,150,165]. However, in the traditional method, handcraft descriptor (Gabor wavelet, Statistical feature, BSIF, SIFT, etc.) are used to obtain the attribute from images [32,91,136,137,152]. Moreover, deep learning can reduce the computation time of biometric authentication. ...
Biometric deals with the verification and identification of a person based on behavioural and physiological traits. This article presents recent advances in physiological-based biometric multimodalities, where we focused on finger vein, palm vein, fingerprint, face, lips, iris, and retina-based processing methods. The authors also evaluated the architecture, operational mode, and performance metrics of biometric technology. In this article, the authors summarize and study various traditional and deep learning-based physiological-based biometric modalities. An extensive review of biometric steps of multiple modalities by using different levels such as preprocessing, feature extraction, and classification, are presented in detail. Challenges and future trends of existing conventional and deep learning approaches are explained in detail to help the researcher. Moreover, traditional and deep learning methods of various physiological-based biometric systems are roughly analyzed to evaluate them. The comparison result and discussion section of this article indicate that there is still a need to develop a robust physiological-based method to advance and improve the performance of the biometric system.
... Afterwards, based on the dogman model, various traditional iris recognition system was designed. The traditional iris recognition techniques are broadly classified into three folds: pre-processing (iris segmentation) [11,96,104], feature extraction [32,73,78,150,152,156 and matching/classification [6,64,98,99,127,[209][210][211]. Iris recognition system affected by the images taken from the different sensor device. ...
... However, the method was mathematically complex. To improve storage requirement and authentication speed, Rathgeb et al. [152] proposed an improved SIFT based Iris recognition system. However, the proposed method is a computational complex. ...
... A deep learning-based method is robust to acquire feature precisely from the raw image, without the involvement of a handcrafted feature, which significantly enhances the matching method [41,60,150,165]. However, in the traditional method, handcraft descriptor (Gabor wavelet, Statistical feature, BSIF, SIFT, etc.) are used to obtain the attribute from images [32,91,136,137,152]. Moreover, deep learning can reduce the computation time of biometric authentication. ...
Biometric deals with the verification and identification of a person based on behavioural and physiological traits. This article
presents recent advances in physiological-based biometric multimodalities, where we focused on finger vein, palm vein, fingerprint,
face, lips, iris, and retina-based processing methods. The authors also evaluated the architecture, operational mode,
and performance metrics of biometric technology. In this article, the authors summarize and study various traditional and
deep learning-based physiological-based biometric modalities. An extensive review of biometric steps of multiple modalities
by using different levels such as preprocessing, feature extraction, and classification, are presented in detail. Challenges and
future trends of existing conventional and deep learning approaches are explained in detail to help the researcher. Moreover,
traditional and deep learning methods of various physiological-based biometric systems are roughly analyzed to evaluate
them. The comparison result and discussion section of this article indicate that there is still a need to develop a robust
physiological-based method to advance and improve the performance of the biometric system.
... Afterwards, based on the dogman model, various traditional iris recognition system was designed. The traditional iris recognition techniques are broadly classified into three folds: pre-processing (iris segmentation) [11,96,104], feature extraction [32,73,78,150,152,156 and matching/classification [6,64,98,99,127,[209][210][211]. Iris recognition system affected by the images taken from the different sensor device. ...
... However, the method was mathematically complex. To improve storage requirement and authentication speed, Rathgeb et al. [152] proposed an improved SIFT based Iris recognition system. However, the proposed method is a computational complex. ...
... A deep learning-based method is robust to acquire feature precisely from the raw image, without the involvement of a handcrafted feature, which significantly enhances the matching method [41,60,150,165]. However, in the traditional method, handcraft descriptor (Gabor wavelet, Statistical feature, BSIF, SIFT, etc.) are used to obtain the attribute from images [32,91,136,137,152]. Moreover, deep learning can reduce the computation time of biometric authentication. ...
Biometric deals with the verification and identification of a person based on behavioural and physiological traits. This article
presents recent advances in physiological-based biometric multimodalities, where we focused on finger vein, palm vein, fingerprint, face, lips, iris, and retina-based processing methods. The authors also evaluated the architecture, operational mode,
and performance metrics of biometric technology. In this article, the authors summarize and study various traditional and
deep learning-based physiological-based biometric modalities. An extensive review of biometric steps of multiple modalities
by using different levels such as preprocessing, feature extraction, and classification, are presented in detail. Challenges and
future trends of existing conventional and deep learning approaches are explained in detail to help the researcher. Moreover,
traditional and deep learning methods of various physiological-based biometric systems are roughly analyzed to evaluate them. The comparison result and discussion section of this article indicate that there is still a need to develop a robust
physiological-based method to advance and improve the performance of the biometric system.
... Hard work and physical activity in some occupations cause people to lose their fingerprints, and identifying them with these methods is challenging [9]. In some cases, people intentionally destroy their fingerprints temporarily or permanently with acid or burns to make them difficult to identify in criminal activities [10]. ...
In this research, biometric authentication, which has been widely used for different purposes in the last quarter-century, is studied. Dorsal hand veins are used for biometric authentication. “Deep learning” (DL) and “generative adversarial networks” (GANs) are used together as keys in the study. A DL-GAN is obtained by combining deep learning and GAN. The developed DL-GAN method is tested on two separate databases. The adversarial network (DL-GAN) method is developed to increase the authentication process’s proportional value. For identity verification, dorsal hand veins with biometric physical properties are used. A multistep approach is used for selecting hand dorsal features, including preimage processing and effectively identifying individuals. The deep learning productive antinetwork method is used to effectively identify individuals based on the information obtained from the dorsal hand vein images. For the test in the study, two open access databases are used. These databases are the Jilin University - dorsal hand vein database and the 11K hands database. The results of the experiments performed on the dataset related to the dorsal hand vessels show that the DL-GAN method reaches an identity accuracy level of 98.36% and has an error rate of 2.47% and a standard accuracy of 0.19%. The accuracy of the experimental results in the second dataset is 96.43%, the equal error rate is 3.55% and the standard accuracy is 0.21%. The improved DL-GAN method obtains better results than physical biometric methods such as LBP, LPQ, GABOR, FGM, BGM and SIFT.
... (Ali et al., 2016) has used SURF for keypoint detection with many different feature matching techniques including Contrastlimited adaptive histogram equalization (CLAHE), histogram equalization (HE) and adaptive histogram equalization (AHE) at different levels for finding which best fits with SURF and enhances iris image recognition. (Rathgeb et al. 2019) has discussed the advantages and significance of using SIFT and SURF descriptors on iris recognition. ...
Iris recognition is a well-known accurate biometric technology and major research area in pattern recognition and computer vision available today. It targets human recognition through the person's iris recognition without human intervention. In many areas iris recognition plays well such as bioinformatics, machine vision, pattern recognition, etc., and it is one of the popular subjects still. Finding of features to identify an iris, which is a small black part of an eye, is a difficult problem in iris recognition. Many methods and algorithms have been proposed on feature extraction, which include aspects like statistical features, level of invariance and robustness. In this article, a traditional SURF and SIFT algorithms are tested for iris recognition. To improve the performance of these algorithms, we passed the input through different domains from the real time. Through applying the Gabor Wavelet Transform (GWT) or Discrete Wavelet Transform (DWT)to the input iris images, a denser and more clear images obtained compared to those by the traditional SURF and SIFT. Thus the simulations of the proposed approaches of using Gabor Wavelet Transform or Discrete Wavelet Transform on SURF and SIFT algorithms gives better results compared to the traditional algorithms.
... Among the keypoint detectors SIFT [2] and SURF [3] are one of the most popular methods. In [4] the authors provide a very good survey of SIFT based methods used for iris recognition. ...
The object of interest of this paper is an original method based on SURF descriptors and texture features applied on iris recognition. Our approach uses SURF method in two stages. In the first step, few keypoints are generated and then combining these results with texture based results allows us perform the second step. In the second stage, one applies again SURF, this time with much more descriptors, thus improving the recognition rate. In our experiments we have tested the influence on the recognition rate of different parameters involved in both the SURF based feature extraction process and the keypoint matching scheme. For tests we employed two iris datasets, namely UPOL and UBIRIS. We use Local Binary Patterns (LBP) and Dual Tree Complex Wavelet Transform (DTCWT) for texture characterization. We experimentally prove that the proposed approach improves the results obtained by applying these methods separately. We show that our method can be successfully used as a filter that reduces the search space for image recognition, the output of this filtering process can be then processed using computationally expensive methods.
... The multimodal system has ability to overcome this limitation by combining evidences from both iris and solves the problem of non-universality. Convolutional neural network has automated the feature extraction process without any domain knowledge and also given outstanding results over Texture code [3] , Discrete Wavelet transform (DWT) [4], Scattering Transform [5] , Shift Invariant Feature Transform (SIFT) [6] and Discrete Cosine Transform (DCT) [7]. Now a much attention is focusses in designing CNN model for automatic feature extraction and suitable classifier for classification purpose [8]. ...
... The Euclidean distance classifier is used for classification. Rathgeb et al. [6] used general image descriptor as SIFT( shift invariant feature transform ) for feature extraction. This work mainly focuses on recognition accuracy of iris recognition. ...
Iris is most promising bio-metric trait for identification or authentication. Iris consists of patterns that are unique and highly random in nature .The discriminative property of iris pattern has attracted many researchers attention. The unimodal system, which uses only one bio-metric trait, suffers from limitation such as inter-class variation, intra-class variation and non-universality. The multi-modal bio-metric system has ability to overcome these drawbacks by fusing multiple biometric traits. In this paper, a multi-modal iris recognition system is proposed. The features are extracted using convolutional neural network and softmax classifier is used for multi-class classification. Finally, rank level fusion method is used to fuse right and left iris in order to improve the confidence level of identification. This method is tested on two data sets namely IITD and CASIA-Iris-V3.
